A Hybrid Random Forest and Least Squares Support Vector Machine Model Based on Particle Swarm Optimization Algorithm for Slope Stability Prediction: A Case Study in Sichuan–Tibet Highway, China

Table 3

Comparison of the performances of the different algorithms on the training and testing sets.

Dataset	Model	Model parameters		Accuracy (%)	F-measure (%)	AUC
		c	σ
Training	GA‒SVM	0.1	0.01	93.74	94.12	0.937
	PSO‒SVM	0.1	14.46	95.12	94.85	0.941
	PSO‒LSSVM	4.77	93.74	100	100	1.00
	RF‒PSO‒LSSVM	2.48	4.89	100	100	1.00
Testing	GA‒SVM	0.1	0.01	70.83	75.86	0.865
	PSO‒SVM	0.1	14.46	83.33	86.67	0.887
	PSO‒LSSVM	4.77	93.74	87.50	89.66	0.944
	RF‒PSO‒LSSVM	2.48	4.89	95.82	96.55	0.964